Table 3.

Local staff survey results according to Promoting Action on Research Implementation in Health Services constructs.

PARIHSa construct	Survey question	Staff in agreement with statement (n=10), n (%)	Linkages to Table 4 theme and whether survey responses are consistent with Table 4 themeb
1. Evidence—caregiver experiences	BBCc has been well-accepted by VAd caregivers	7 (70)	Theme 2a: Yese
2. Evidence—research evidence	BBC is supported by research evidence	7 (70)	Theme 2b: Yes
3. Context—staff resources	Informational materials are available for caregivers and staff to raise awareness about the workshop	7 (70)	Theme 3b: Yesf Theme 4d: Nof
4. Facilitation—staff skills and attributes	Resources to educate staff about workshop structure and content are available	7 (70)	Theme 5a: Partialg
5. Facilitation—availability of external facilitator to mentor and assist staff	A regional mentor who can help answer questions or solve problems is available	3 (30)	Theme 5a: Yes
6. Context—information technology capabilities	An approach used at our facility to evaluate and improve implementation includes a system of tracking which caregivers have been referred to BBC	4 (40)	Theme 5b: Yes

^aPARIHS: Promoting Action on Research Implementation in Health Services.

^bLinkages (including areas of consistency and inconsistency) between quantitative and qualitative results were discussed during our process of integrating stage 1 findings, as described in the Methods section.

^cBBC: Building Better Caregivers.

^dVA: US Department of Veterans Affairs.

^eCaregiver responses to survey questions on the PARIHS evidence construct are described in the text and are consistent with these staff responses, as well as with Theme 2a.

^fExplanation for both a yes and a no designation: qualitative results in the text and Table 4 indicate that certain outreach materials, particularly those useful for post-9/11 caregivers, were readily available (and used by staff) during workshop rollout, which is consistent with Theme 3b. However, outreach materials, mechanisms, and contacts for pre-9/11 caregivers and others were insufficient (as summarized by Theme 4d), which explains why this survey finding was felt to be inconsistent with Theme 4d.

^gExplanation for partial designation: qualitative results in the text and Table 4, Theme 5a, explicitly recognize that training was insufficient for many new staff members but, conversely, they also imply that experienced staff (present during the early rollout) were generally satisfied with their training, which explains why this survey finding was determined to be partially consistent with Theme 5a.